IEEE ©1997 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

Return to Table of Contents

A Unique Desk-top Electrical Machinery Laboratory for the Mechatronics Age

Tatsuya Kikuchi, Member, IEEE, and Takashi Kenjo, Member, IEEE


Abstract - This paper describes an electrical machinery laboratory, named MECHATRO LAB 2PLUS, designed to cope with modern realities in which engineering graduates encounter various kinds of servomotors and actuators much more frequently than conventional power electric machines. The set-up can demonstrate the principles of some 20 different kinds of electric machines ranging from conventional rotary machines to modern control-use motors such as stepping motors, brushless motors, and switched-reluctance drives. The experimental themes number more than 100, including elementary subjects, microprocessor-controlled power electronic drives, and control/drive programming using various computer languages. Unlike conventional motor-generator workshop equipment, a remarkable feature of our desk-top bench is that it does not occupy much space on a permanent basis. Along with the background of its design history, which covers more than 20 years, the uniqueness and advantages of the new experimental set-up are presented in detail, citing some sample experiments and reactions of Japanese students and overseas instructors.



I. Introduction

Today's high-tech industrial society is supported by massive numbers of small precision motors, used in robots on the factory floor, CNC machines, storage devices, printers, digital videodiscs (DVDs), etc. A luxury car, for instance, employs nearly 100 motors to provide for passenger comfort. These motors are controlled by advanced digital technology combined with power electronics. Industry thus demands a supply of technically trained manpower that can meet such advanced technologies.

However, in most university engineering departments, the course on electrical machines deals mainly with magnetic circuits, transformers, and large induction or synchronous motors. While these power machines should not be ignored in the electrical engineering curriculum, some amount of time should be allotted to small control-use machines. In reality, however, it is difficult to provide sufficient time for small machines since new subjects, such as computer hardware/software or engineering ethics, have been added to the curriculum. This is an issue that cannot be solved at the level of individual instructors.

A problem also exists on the students' side. As pointed by Nehrir et al. [1], students generally feel that electrical machines are not high-tech equipment and so are not interested in learning them. (Our university is no exception.) There have been various attempts to make the course more attractive. For example, Brice [2] designed a set-up where traditional analog measuring devices are replaced with the latest PCs, and proposed a workshop subject that unifies electrical machinery, signal processing and software engineering.

With regard to the classroom environment, Lubkeman and Collins [3] presented a method of making up for the intrinsic deficiencies of conventional classrooms, i.e. hypermedia-based courseware aimed at cultivating an intuitive understanding of theory and practice.

These studies all target conventional machines, however, and to the authors' knowledge, there are only a few reports dealing with the educational problems regarding small machines. For students majoring in electronics or information science, it may be reasonable to exclude large power electric machines from their curriculum. But small motors and their controls are essential subjects that should be studied in universities, not only because these technologies form an important component within Japanese industry, but also because the motion of computer peripherals is governed by small motors/actuators.

The authors have for some time been engaged in research and education related to small motors, designing in the process several experiment benches and control instruments to be used in normal classrooms, workshops or R&D laboratories. In this paper, we describe how the final design of the desk-top electrical machinery experiment bench, which we name MECHATRO LAB, was reached, citing some specifications, construction outlines, typical experiments and their features.


II. Development Background

For the first stage, we designed in 1971 various 200W-class AC motors. The rotors are shown in Fig. 1. Two key points were realized here: downsizing (miniaturization) and the inclusion of special motors such a hysteresis motor, a solid-steel eddy current motor, a surface-conductor motor and a reluctance motor. The brushless DC motor, in those days known as a commutatorless motor, was constructed by a student as a final-year project and later included in the set in 1974.


Fig. 1. Rotors of the first prototype.

With the advent of the microprocessor in 1971 [4], digital electronics technology has become a key component in the field of mechatronics. Meanwhile, the small precision motor industry has been making spectacular progress in Japan since the late 1970s. In 1981, therefore, we began designing a single-board microcomputer, named KENTAC, for mechatronics education purposes, using the 8080, 8085, Z80 and 8086 series of microprocessors. Based on this, we then decided to design a training bench combining electrical machines and microprocessor-controlled power electronics. After a few trials, we put together in 1984 the first generation of MECHATRO LAB, which is referred to as ML1 and shown in Fig. 2. Details of this machine are given in reference [5]. Figure 3 shows its basic system. The motors employed in this model are all small: the stator and rotor diameters are 80 mm and 39.5 mm, respectively. As seen in Fig. 4, a key feature of this device is the inclusion of a variety of rotor constructions. Portability, compactness and easy insertion and removal of rotors were important features as well. A single capacitor was used for generating three-phase current from a single-phase 50/60 Hz main (this technique is not suited to larger motors). The rotors are supported at one end of the shaft for easy insertion and removal, but load testing was not available. More than 300 sets were placed in technical colleges and schools, vocational training facilities, and universities around Japan. MECHATRO LAB 1 is being used also in several vocational training centers outside of Japan.


Fig. 2. Overall view of MECHATRO LAB 1.


Fig. 3. System diagram of MECHATRO LAB 1.


Fig. 4. Construction of various rotors.

Early responses from instructors using ML1 included requests to incorporate loading tests and to make it possible to combine two machines for demonstrating a motor-generator set. To meet these requests and upgrade its utility, we developed in 1993 a new experiment bench, MECHATRO LAB 2 (ML2), shown in Fig. 5. A major difference from the first version is the machine size: the average rotor volume is almost three times that of ML1. This enlargment was needed for load testing. In addition, small motors require expensive, fine measurement equipment, so a separate measurement unit was added. The version we present here is not the commercial version, but the latest laboratory design incorporating major refinements. We refer to this as MECHATRO LAB 2PLUS (ML2PLUS). With this bench, conventional machines and various small motors can be demonstrated. Students can learn visibly the construction and material makeup of some 20 different electric machines, using 12 different types of rotors and four stators. Moreover, using a power MOSFET circuit and a computer, students can learn interdisciplinary, high-level programming for motor drives. With the use of the measurement unit and computers, over 100 different experiments are possible, ranging from those on basic drives, torque tests, to speed tests.


Fig. 5. Overall view of MECHATRO LAB 2; the stators seen are a distributed winding type, a six-coil concentrated-winding type and a two-pole permanent-magnet type.


III. Construction of the Latest Version MECHATRO LAB 2PLUS

MECHATRO LAB 2PLUS consists of four units as shown in Fig. 6: the electric machinery unit, the control unit, the measurement unit, and the power supply unit. Each unit is outlined below.


Fig. 6. System diagram of MECHATRO LAB 2PLUS.


A. Electrical Machinery Unit

We have provided the following four different stators:

1) Stator A: Distributed windings for four- and eight-pole schemes using a 24-slot core (see Fig. 7). There are six terminals for each scheme on the panel, for which either a delta or star connection is possible using banana jacks with wires.


Fig. 7. Stator A: Distributed winding type.

2) Stator B: Six-coil concentrated winding type (see Fig. 8). Twelve terminals forming six coils appear on the panel. When they are connected so as to generate different magnetic polarities on any two opposing poles, the stator will work in a two-pole scheme. If connected to produce the same polarity, the stator becomes either four- or eight-pole depending on whether a four- or eight-pole rotor is inserted. This multipurpose stator can be used for induction, synchronous, stepping, brushless DC, switched-reluctance, universal and wound-DC motors.


Fig. 8. Stator B: Six-coil concentrated winding type.

3) Stator C: Inductor-pole stator (see Fig. 9). Fine teeth, or inductors, are cut on the six poles with concentrated windings. This stator is used for the hybrid stepping motor, and the same set-up can be used to demonstrate the principle of the inductor generator.


Fig. 9. Stator C: Inductor-pole stator.

4) Stator D: Permanent-magnet field stator (see Fig. 10). Ferrite magnets form a two-pole configuration. This stator is used exclusively for a conventional DC machine.


Fig. 10. Stator D: Permanent-magnet field stator.

Rotors: 12 rotors are provided:

1) Squirrel-cage rotor with copper bars and end rings (see Fig. 11). This rotor is used for an induction machine.


Fig. 11. Squirrel-cage rotor with copper bars and end rings.

2) Squirrel-cage rotor with brass bars and end ring (see Fig. 12). This rotor is also used for an induction machine. However, the starting torque is higher than that of the copper rotor.


Fig. 12. Squirrel-cage rotor with brass bars and end rings.

3) Salient-poled squirrel-cage rotor (see Fig. 13). This rotor is used exclusively for the reluctance-type synchronous motor.


Fig. 13. Salient-poled squirrel-cage rotor.

4) Semihard steel rotor (see Fig. 14). The main material is an alloy of iron, aluminum and nickel. This rotor is used for the hysteresis synchronous motor, which is an AC motor utilizing magnetic hysteresis for torque generation. Interestingly, the rotor can run stably at a synchronous speed and also at lower speeds, indicating that the hysteresis motor is a synchronous motor that also can be an asynchronous motor.


Fig. 14. Semihard steel rotor.

5) Solid-steel rotor (see Fig. 15). The rotor's main part is a ring of mild carbon steel. This rotor is for the eddy-current motor, a special-type of induction motor.


Fig. 15. Solid-steel rotor.

6) Salient-poled laminated silicon-steel rotor (see Fig. 16). This rotor is for a variable reluctance (VR) stepping motor.


Fig. 16. Salient-poled laminated silicon-steel rotor.

7) Rotor for use with the inductor-pole stator (see Fig. 17). A ferrite ring magnet is sandwiched by two stacks of lamination core with 20 teeth on their periphery. This rotor-stator combination produces a hybrid stepping motor, but when the rotor is driven by an external force, it can be a high-frequency inductor generator.


Fig. 17. Rotor for hybrid-type stepping motor.

8) Two-pole permanent-magnet rotor (see Fig. 18). This rotor is used for demonstrating a primitive synchronous motor or the principle of the permanent-magnet stepping motor. Stator B is used in the two-pole scheme.


Fig. 18. 2-pole permanent-magnet rotor.

9) Four-pole permanent-magnet rotor (see Fig. 19). Combined with stator A or B, this rotor is used mainly for demonstrating a brushless DC motor. With stator A, it can also show the operation of a permanent-magnet synchronous motor.

Fig. 19. 4-pole permanent-magnet rotor.

10) Eight-pole permanent-magnet rotor (see Fig. 20). When combined with stator B, this rotor becomes an eight-pole brushless DC motor. Interestingly, this rotor and the four-pole one are interchangeable when combined with stator B. In real designs, both cases are seen.

Fig. 20. 8-pole permanent-magnet rotor.

11) Commutator rotor (see Fig. 21). This rotor is used for either a conventional DC or AC series motor (i.e., universal motor).


Fig. 21. Commutator rotor.

12) Wound-field rotor (see Fig. 22). This rotor has three applications: wound-rotor induction motor, synchronous motor and synchro/resolver.


Fig. 22. Wound-field rotor.


B. Control Unit

This unit comprises a set of manual switches, a MOSFET power electronics circuit, an LC power filter, and a microcomputer board (see Fig. 23). Details are as follows:

1) The six manual switches, which were also used in ML1, can be used, for example, for demonstrating the principle of a three-phase inverter. By manual operation, direct voltage/current can be converted to three-phase alternating voltage/current, although they are not sinusoidal. It is important to show students that non-sinusoidal, rectangular alternating voltage can drive an AC motor and is in fact suited to a stepping motor. This set-up is often much better than the complicated transistor/MOSFET inverter in practical use when students are trying to understand the principles of the inverter or for checking the switching sequence. Thus, manual operation of these switches can sufficiently demonstrate the essential principles of digital operation of an electric motor.

2) The power electronics circuit consists of six power MOSFETs. It is a universal or multipurpose circuit, which can be used even for a real industrial motor. The switching signals are supplied by the Z80 of the control unit or an external PC. Two modes are provided: one for direct interface with a PC, and the other via a specially designed interface board.

3) The LC filter is for providing sinusoidal voltages from pulse-width-modulated (PWM) three-phase power. In ML1, the single-phase AC was converted to "quasi-three-phase" AC with a single capacitor. However, in ML2, perfect three-phase is provided by this method. The filter is also used for comparison between sinusoidal operation and PWM drive.

4) The ROM in the control unit memorizes eight programs (six-step inverter, PWM inverter, single-phase operation of three-phase stepping motor, brushless DC motor, etc.) The CPU used here is a Z-80 operated at a clock frequency of 8MHz.


Fig. 23. Control unit.


C. Measurement Unit

The measurement unit compactly houses eight kinds of measurement instruments needed for motor tests: DC voltmeter, DC ammeter, AC voltmeter, AC ammeter, three-phase wattmeter, frequency meter, tachometer, and torque meter. By wiring using banana jacks and plugs, various measurements can be made with ease.


D. Power Supply Unit

Two power supplies are needed when conducting experiments with the separately-excited DC motor. The two supply systems (18V/2A and 60V/4A) are both provided with short-circuit protection.


E. Computer

The microprocessor on the drive board does not have functions which allow for running students' software. Instead, two options exist for programming practice: either a normal (IBM/NEC) personal computer or a specially designed controller, named KENTAC mark2 or KENTAC rm86, is used. In the 1980s, we mainly used KENTAC mk2, which incorporates an 8-bit Z80, and sometimes KENTAC rm86 (16-bit 8086-type CPU) with assembly language. In the 1990s, we have employed C language on a regular PC, in line with industry's current trend of using C.


IV. Experiment Organization

Various motor types are used in industry, and each has its own design concept, which results in different sizes, winding types, and construction. Some good textbooks on the subject provide well-illustrated materials on motors that can provide students with a clear understanding, but the interrelationship among the different types is usually not explained. In school education (including university), it is often more important for the student to grasp the relationship existing between two or more apparently different things, rather than to get bogged down in specific details. For instance, instead of learning details of a stepping motor and brushless motor as separate items, making students understand the similarities and differences of the two motor types is of more educational benefit, particularly when our primary goal is to cultivate innovative minds.


A. Twenty-two Rotating Machines

In upgrading the design of MECHATRO LAB, we kept the electronic equipment unchanged, but attempted a more thoroughly systematic coverage of electrical motors so that instructors could demonstrate the principles of important motors of the past and those in present use. In particular, special attention was paid so that models of historical or epoch-making machines could be reconstructed with ML2, and a survey of such machines could be introduced into the course.

The small two-phase stepping motor mounted on ML1 was removed for ML2, since we thought that Stator B (six-coil concentrated winding type) combined with an appropriate rotor suffices for constructing variable-reluctance and permanent-magnet stepping motors. The experimental scope of the ML2PLUS version was then enlarged by the addition of an inductor-pole stator, a rotor for a hybrid stepping motor and a wound-rotor having slip-rings.

Figure 24 shows 22 different types of rotating machines that can be demonstrated on ML2PLUS. Although there are other motor types, such as the moving-coil motor, printed or pan-cake motor, and ultrasonic-wave motor, these special types are not included in the bench. When they are needed for classroom demonstration, we show the actual motors or a model designed as a teaching aid.


Fig. 24. Twenty-two rotating machines that can be demonstrated in ML2PLUS.


B. ML2PLUS Organization Tree for Major Experiments

Figure 25 shows an organization tree for experiments possible with the above machines. In addition to ordinary experimental topics such as torque-vs.-speed characteristics, experiments demonstrating the interrelations between motors, power electronics and computers are included (for instance on the brushless DC motor or micro-stepping drive of a stepping motor). The experiments cover motors as well as generators. Motor experiments include those on speed controls, loaded characteristics, manual inverters, and programming tests, while generator experiments consist of the five topics shown by the classification numbers in Fig. 25.

The relation between a generator and motor can be demonstrated, for example, using the high frequency generator and the hybrid stepping motor. When the rotor shown in Fig. 17 is rotated in Stator C (of Fig. 9 ), an alternating voltage waveform is generated, whose frequency is much higher than the rotational speed, e.g. 1kHz at the speed of 50 revolutions per second. We explain to students that the term 'hybrid' refers to the combined effects of using a permanent magnet (for high efficiency) and fine teeth on the rotor periphery (to obtain small step size). The hybrid stepping motor appeared after the invention of this generator, and advancements in electronics and computer were essential for the proliferation of stepping motors.


Fig. 25. Organization tree for major experiments using ML2PLUS.


C. Nonlinearity Problems

The effects of nonlinear magnetic properties of the materials used in an electric machine are important in many ways. In the experiment on the shunt DC motor using ML2, the student can see the effect of armature reaction, i.e., the overall reduction of the field flux due to armature current. The permanent magnet DC motor, on the other hand, is not subjected to armature reaction because of the high coercive force of the ferrite magnet. A more general question is, "at what stage should we introduce topics such as nonlinearity?" The textbook we use for sophomores presents a linear theory for the principles of stepping motors. We explain that such a simple theory does not always hold in a real motor where deep magnetic saturation is occurring, and recommend further reading of the Japanese version of Chapter 4 of reference [6], which gives a torque theory for reversible magnetic nonlinearity.

The principle of creating a high torque using magnetic hysteresis is discussed in detail in our sophomores' textbook and briefly summarized in Chapter 3 of reference [7]. Regardless of whether MECHATRO LAB is used or not, in classes where the students' level is high, we discuss this subject, but skip it if the average student does not seem to have enough background. For those who show a strong interest in magnetic hysteresis regarding motor design, we recommend reference [8], which gives a mathematical treatment of irreversible nonlinearity, and explains the apparent differences and similarities between hysteresis and ultrasonic motors.


V. Sample Experiments

ML2PLUS is capable of demonstrating more than 100 experiments, summarized by the experiment tree, on some 22 types of electrical machines, ranging from classic experiments for motors and generators to modern computer-based controls. However, it does not cover all topics required for training designers of electrical machines. In particular, power engineering majors are required to take a workshop course on larger machines during their junior year. Below we present some typical topics using ML2PLUS, with reference to the classification tables, and remark on some of the unique features of the experimental set-up.


A. Assembling Motors and Generators

This experiment concerns the basic practice of assembling a specific motor type by combining the correct rotor and stator and making the proper wiring connections. The brushless DC motor provides a good example for learning how to organize a machine system. Although the student may select any of the three permanent-magnet rotors, we first recommend the four-pole rotor. There are two choices for the stator: stator A with its four-pole windings, and stator B (six-pole stator) in the four/eight winding connection. As the position sensor, a Hall-element assembly and a disk magnet, illustrated in Fig. 26, are provided; the disk magnet is mounted on the rotor shaft while the Hall-element assembly is fastened to the bracket (i.e., end bell). One assignment is to write a program in Z80 assembly language or C for generating switching signals to be fed to the inverter, in reference to position signals from the Hall-element sensor. The instructor can use the software stored in the ROM. To achieve a perfect drive, the student must consider the relations between rotor position, sensor signal timing, and inverter programs.


Fig. 26. Position sensors using Hall-effect elements.


B. Velocity Control Laboratory

In motion control, or mechatronics, various techniques are used to control or adjust speed. In ML2PLUS, at least 17 principles can be demonstrated. A classic example is given by field flux adjustment in a separately-excited DC motor (see Fig. 27; two opposing poles are used as the field poles here, but it is also possible to use four or all six poles).

This experiment can be extended to a topic where the wire size and number of turns for the field system are compared with other excitation schemes such as in series or shunt motors. Actual series motors have thick field coils and a relatively small number of turns, while a shunt motor has a thin wire turned many times. A student assignment can be to discuss the reasons for this difference, and to figure out how to employ six coils to create the same effect.


Fig. 27. Separately-excited DC motor using the 6-pole stator.


C. Manual Switching Laboratory

Many fundamental experiments can be carried out using the manual switches. Examples include reversing the DC motor's rotational direction, and inverter drive of synchronous, induction, variable reluctance (VR) and hybrid stepping motors. The VR motor employs a rotor that has a core with a cross-shaped cross-section (i.e., a four-pole salient type), together with stator B (with six poles). Important items to note in this experiment are: 1) the principle of traveling one step angle by a single switching operation, a fundamental principle in digital control; 2) that the rotational direction is determined by the switching sequence or by changing the wiring connection; and 3) settling behavior. While the simple single-phase-on movement generates long-lasting oscillations, two-phase operation displays good damping that rapidly settles. For this laboratory, we show a variable reluctance stepping motor that was used in a British warship in the 1920s[7] (Fig. 28), and note that 1) the problem of oscillation already existed at that time, and 2) the means of switching has progressed from mechanical to electronic today.


Fig. 28. Citing a historical motor: A variable-reluctance motor.


D. Observation of Back Electromotive Force

The back-emf waveforms are important factors when designing brushless DC motors. Simple but highly instructive experiments are available using the three permanent-magnet rotors and stators A and B. Figure 29 shows examples of these waveforms. We demonstrate that different waveforms are produced by delta and star connections, and also by using single-coil and star/delta.

In addition, by using the MOSFET inverter, we can show how a three-phase electromotive force is rectified by the diodes built in each MOSFET. An instructive question that can be posed to the student is: in a DC machine composed of a commutator rotor and stator D (permanent-magnet field stator), why does the back-emf, as observed across the two terminals, reverse its polarity when the rotor's motional direction is reversed, while the rectified back-emf generated by stator A or B does not? The explanation given to the student is that the commutator and brushes work to convert the alternating electromotive forces to dc voltage. The three-phase rectifier also converts AC to DC. Thus, the DC motor rectifies automatically by referring to the coil positions, while the diode bridge carries out rectification due to the diode's property, independent of the rotor position. This is another example of demonstrating similarities and differences.


Fig. 29. Showing differences in back-emf when the four-pole permanent magnet rotor is rotated in stator B (observed at different speeds). (a) Star connection. (b) Delta connection.


E. Load Characteristics Laboratory

Torque measurement is carried out by use of a pulley, weight, and pressure sensor using the piezoelectric effect, as shown in Fig. 30.

T = rg( W - w )
where
r = pulley radius
W = mass of weight
w = pressure sensor reading
g = acceleration of gravity

Using this principle and the speed sensors equipped, students can measure the torque-vs.-speed characteristics of a DC motor and an induction motor.


Fig. 30. Torque testing using a pulley, weight and pressure sensor.

Another method of torque measurement employs pulleys and a belt connected with the permanent-magnet DC motor. The permanent magnet DC motor can be used as a brake and torque meter. By adjusting the applied voltage, we can control the brake torque (which is determined from the armature current I ).

Here, we can demonstrate a simple but useful property of the permanent-magnet DC motor, namely, that the torque T is given by

T = K (I - J )

where K is the so-called back-emf constant and J the loss current which is mainly related to the mechanical friction between the pulley and belt. Determination of K and J offers a good experiment subject.

While ML2PLUS includes a hysteresis brake, it is not essential because the principle of this brake can be demonstrated using stator A and the semihard steel rotor.


F. Digital Three-Phase Quantities Interpreted in Three-Dimensional Space

The relationship between ordinary sinusoidal three-phase voltage and the three-phase voltage generated by a combination of digital techniques and switching provides an interesting experimental subject. When a physical or mathematical problem is examined in three-dimensional space, three orthogonal axes are usually used. However, students studying electric power are told that three-phase power, with phases shifted 120 degrees to one another, is not only fundamental but very convenient in various ways. Inquiring students may wonder whether these two concepts are related. As a clue linking orthogonality and 120-degree shift, we present Fig. 31, where each corner of a cube represents a digital switching state of the three-phase bridge circuit of an inverter. When the three orthogonal axes are viewed at the angle connecting the origin (0,0,0) and the corner (1,1,1), the resulting two-dimensional image is a hexagon, with the three axes 120 degrees apart. Based on a mathematical analysis of this model, it is possible to produce some peculiar series of PWM switching signals that generate sinusoidal PWM signals across any two phases. An example is shown in Fig. 32: although the switching signal for each phase seems unrelated to a sine wave, the difference between two phases is a pulse-width modulated sine wave. Moreover, the maximum voltage attained by this method can be higher by a factor of 2/SQR3than that with normal sine-wave modulation carried out for each phase. (Details are discussed in reference [5].) The software for generating the switching signal shown in Fig. 32 was written by a student, using the inverter in ML1 and a KENTAC mark2, as a final year project before ML 2 appeared.


Fig. 31. Intuitive understanding of relationship between three-dimensional orthogonal axes and 120-degree-shifted three-phase.


Fig. 32. Non-sinusoidal PWM signals to produce sinusoidal PWM current.


G. Synchro and Resolver

Figure 33 shows a set-up for demonstrating the synchro using two wound-rotor machines. The stator terminals of the two machines are connected to each other and a single-phase current is supplied to both rotor terminals (to only one phase each). When the shaft of one machine is rotated, the other automatically travels the same angle.

To demonstrate a rotary transformer and resolver we use one machine. When 50/60Hz single-phase current is supplied to a phase winding of the stator, the induced voltage in the rotor's two phases varies with rotor position. The resolver detects the rotor's angular position from a comparison of magnitude and the phase angle (0 or 180 degrees). The measurement of magnitude and phase as a function of angular position provides a good experimental topic. We can also use the wound-rotor to demonstrate a wound-rotor induction motor, although our small model cannot display the ideal torque-vs.-speed characteristics given in textbooks.


Fig. 33. Synchro transmitter and receiver.


VI. Discussion

The various versions of MECHATRO LAB have been used in several different courses at our university. In 1996 when we were developing the latest version (ML2PLUS) we carried out an investigation of its effectiveness through questionnaires to Japanese sophomores (in Japanese) and overseas instructors (in English). The following are their reactions along with our comments.


A. Undergraduate Response to Class Utilizing Multimedia together with ML2PLUS

In our latest class for 30 sophomores we incorporated multimedia together with ML2PLUS. We conducted many of the classes in the computer room, using a newly developed HTML document which includes photos, illustrations and animation, in addition to the normal textbook that the author (Kenjo) had been using for over 10 years.

Questionnaire results and exams taken by students showed that a wide range existed in the students' understanding of the subjects treated with MECHATRO LAB. Generally, they were able to comprehend the workings of DC and AC motors more than those of other machines and power circuits such as inverters. This result may be related to the fact that most students have some prior knowledge about brush DC motors, perhaps from having looked at the basic construction of small DC motors as they assembled models or toys when they were younger. Meanwhile, AC rotary machines are rather easy subjects for instructors to teach, since various teaching methods have been contrived for these machines.

On the other hand, students seem to have difficulty with brushless DC motors and stepping motors because they require digital and power electronics for their operation, with which students are unfamiliar even though large numbers of these motors are used in PCs and their peripheral equipment. Even if they had previously looked at a real one, these motors usually cannot be disassembled and inspected, unlike a DC motor for toys. Hence it is difficult for students to understand the principles of these sophisticated motors only from class instruction and textbooks.

In order to determine the effectiveness of ML2PLUS, 30 sophomores were asked to respond to a questionnaire. Some of the positive reactions include:
"Seeing the demonstrations helped me in understanding in a concrete manner."
"In comparison to lectures only, the demonstration made it easier to understand."
"I could directly experience the theory."
"The lab made me become more interested in the subject."

These responses show that, at the very least, instruction using ML2PLUS positively complements traditional teaching methods using a normal textbook.

The students also pointed out some problems, three of which are discussed below:
1) "Too much time is spent by the instructor in making the connections."
Although changing connections between delta and star, or two-pole and four-pole, is not a totally useless exercise, it can be an obstacle when moving on with the lecture/demonstration. To cut the time used for connections, a special harness may be needed.

2) "Demonstrations don't help when one hasn't understood the instructor's explanations."
This comment points out the importance (for the instructor) of concentrating on the essentials rather than elaborating on various permutations.

3) "Difficult to see from the back row."
Improving the classroom layout and utilizing audio-visual aids should solve this problem. Note that using several bench sets is not always effective, especially when the instructor must cover the material within a limited time frame.


B. Response of Overseas Instructors

Very enthusiastic reactions were seen when the latest version (at its development stage) was used in a refresher class for technical instructors from various countries (Mexico, Columbia, Paraguay, China and Malaysia.). This small class was conducted not in a regular classroom but in the author's well-equipped electromechanical laboratory (see Fig. 34), which had on hand a wide range of real motors, tools and instruments and was a much better environment for hands-on learning than the classroom demonstrations for undergraduates cited above.

The most striking feature of this machine, as perceived by them, is the very compact but sophisticated combination of electromagnetic devices (motor/generator), semiconductors (power electronics) and digital controls.

Some of their comments are:
"Today's experiments are really very interesting and I believe we can learn a lot of things from these experiments."
"Today I was learning a lot of subjects that were not very clear before. This practical class is always very useful, because we can see, touch, and make the things."
"Today was a very useful day. I'd like to spend more time in this laboratory, because here you have advanced techniques for dealing with electrical motors, and I'm sure I have a lot to learn about them."


Fig. 34. Overseas instructors discussing around ML2PLUS in the author's laboratory.

Recent overseas trainees at our university from Spanish speaking countries are generally very enthusiastic about learning Japan's latest technologies. This small class showed a keen interest in ML2PLUS's possibilities for studying fine power electronics for motion control of small motors and for surveying historically important technologies which were born in Europe and the United States. In fact, a theme we adopted when undertaking the latest upgrade for ML2PLUS was Confucius' saying " one can learn new things by surveying the past. " What we found out was that " by employing the latest technology, one can review the past effectively."


C. General Discussion

A major issue in finalizing ML2 was to make it so that important devices could be historically reviewed. Professor Y. Kohya of Saga University pointed out that the wound-rotor should be included not only because this provides a model of a large induction motor but also the basis of the rotary transformer and can also demonstrate the principle of the synchro/resolver. We chose a 16-slot core for the rotor to set two-phase, four-pole windings because it can serve as the rotor of an induction motor as well as of a synchro. This approach is another good example of showing a principle common to two apparently different things, in this instance, a motor and a sensor/actuator. The motor principle is realized as a big machine, while the sensor/actuator is a precision device.

One drawback with this model is that a timing belt must be used to couple two machines in the motor-generator experiment. The manufacturer, Showa Dengyosha Co., Ltd. is currently planning a revised model in which two machines can be directly coupled like an actual motor-generator with minimum coupling loss. This arrangement will require some innovative scheme so that the current features of portability and quick removal/insertion of rotors can be retained.

The Ward Leonard system, which is the best position control configuration by classical means, can be demonstrated using ML2PLUS. By showing the essential difference between this approach and modern methods based on sophisticated electronics beyond the level of MECHATRO LAB's electronic equipment, we hoped to inspire in students a respect for the mysterious principles governing electromagnet-mechanical phenomena and make them aware of the deficiencies of semiconductor technology as a future issue to be solved.

As so far described, many experiments are available, several of which may be selected for electronics or information science students by the instructor's judgments. We think that MECHATRO LAB is useful to mechanical engineering students as well.


VII. Conclusions

This paper describes the development of a desk-top electrical machinery experiment bench, including its design philosophy, hardware details and the reactions of students and overseas instructors. The major aim of the bench is to serve as a tool for showing the abundant possibilities of electromechanical actuators, in which motors, semiconductor devices and computer controls are combined. While we have shown that the bench can be effectively employed within university education, it should benefit other facilities such as company training centers as well.

It is the authors' educational philosophy that teaching the history of technology is much more relevant than providing training on current technology, and ML2PLUS serves this purpose successfully.


Acknowledgments

First of all we would like to acknowledge Seishiro Tamura, who helped us design the machine dimensions, and determine winding details, material choices and manufacturing method. He was a young military engineer until the end of World War II, but since then he determined to devote himself to Japan's recovery in the area of small electric motors, although he is not widely known in the field. Only a few people know that he was an important consultant to Sony Corporation when the company was in its infancy, designing the tape-recorder. Not only did he design their hysteresis synchronous motors, but he also produced hundreds of types of motors for use in information equipment manufactured by many firms. When the author (Kenjo) was designing motors at TEAC Corporation, he was bestowed invaluable knowledge by this prominent engineer.

Katsumi Egawa, who in his 30s designed the world's first stepping motor that was launched onto the lunar surface, enthusiastically helped us in prototyping the hybrid stepping motor for upgrading our design (ML2PLUS).

We are also indebted to Hideo Yoshiwara, President of Showa Dengyosha Co., Ltd., for supplying us technical information on their teaching equipment products. He has been enthusiastic about designing the commercial version of MECHATRO LAB to be used in universities, technical colleges, vocational training facilities, company in-plant education centers in Japan and overseas (Turkey, Saudi Arabia, Pakistan, Brazil, Mexico, China, Republic of Korea, and Indonesia).

Finally we would like to give our thanks to R. Takeguchi who kindly brushed up the authors' English when he was very busy as a professional translator before the spring holiday week.


References

[1] M. Hashem Nehrir, Fereshteh Fatehi and Victor Gerez, "Computer Modeling for Enhancing Instruction of Electric Machinery," IEEE Trans. Educ., vol.38, no.2, pp.166-170, May 1995.
[2] Charles W.Brice, III, "Design of a New Electromechanical Systems Instructional Laboratory," IEEE Trans. Power Syst., vol.6, no.2, pp.872-875, May 1991.
[3] David L. Lubkeman and Edward R. Collins, "Hypermedia-based Courseware Development for Power Engineering Education," IEEE Trans. Power Syst., vol.6, no.3, pp.1259-1265, Aug. 1991.
[4] Stanley Mazor, "The History of the Microcomputer - Invention and Evolution," Proc. IEEE, vol.83, no.12, pp.1601-1608, Dec. 1995.
[5] T. Kenjo, Power Electronics for the Microprocessor Age, U.K.: Oxford Univ. Press, 1990.
[6] T. Kenjo and A. Sugawara, Stepping Motors and their Microprocessor Controls, 2nd ed. U.K.: Oxford Univ. Press, 1994.
[7] T. Kenjo, Electric Motors and their Controls, U.K.: Oxford Univ. Press, 1991.
[8] T. Sashida and T. Kenjo, An Introduction to Ultrasonic Motors, U.K.: Oxford Univ. Press, 1993.


Contact Information

Tatsuya Kikuchi
Department of Electrical Engineering
Kanto Polytechnic Center
78, Minamikibougaoka, Asahi, Yokohama, Kanagawa, 241 JAPAN
Phone: 81-45-391-2848
Fax: 81-45-391-0141
E-mail: tkikuchi@uitec.ac.jp

Takashi Kenjo
Department of Electrical Engineering and Power Electronics
Polytechnic University of Japan
4-1-1, Hashimotodai, Sagamihara, Kanagawa 229-11 JAPAN
Phone: 81-427-63-9140
Fax: 81-427-63-9150
E-mail: kenjo@uitec.ac.jp


Biographies

Tatsuya Kikuchi (M'95) received the B.S. degree in Electronic Engineering from the Polytechnic University, Kanagawa, Japan, in 1984. From 1985 to 1991, he was a design engineer of servomotor controls. From 1992 to 1994, he was an Instructor in the Department of Electronic Engineering, the Chubu Polytechnic Center. Since 1995, he has been working at the Kanto Polytechnic Center. His interests include mechatronics and multimedia computing.

Takashi Kenjo (M'97) was born in Japan on February 2, 1940. He received the Masters Degree in 1964 and the Doctor-of-Engineering Degree in 1971 from Tohoku University, Sendai, Japan. His area of interest is in small precision motors and their controls, and he has written several monographs published by Oxford University Press. He has been with the Polytechnic University of Japan since 1965 and is currently a Professor in the Department of Electrical Engineering and Power Electronics.